CS : Spatial Data Modeling and Analysis. Geovisualization

Transcription

1 CS : Spatial Data Modeling and Analysis Geovisualization

2 Visual Perception Learning Styles & Personality Types: Visual, Auditory, Kinesthetic

3 Cholera cases in the London epidemic of 1854

4 Cholera cases in the London epidemic of 1854

5 Cholera cases in the London epidemic of 1854 Broad St. Water Pump

6 Geo-Visualization What is the ratio between areas of Africa and Greenland?

7 Geo-Visualization What is the ratio between areas of Africa and Greenland? 14:1

8 Map Orientation and Projections Mapping a 3D globe on a flat 2D plane 8

9 Map Orientation and Projections Mapping a 3D globe on a flat 2D plane 9

10 Map Orientation and Projections 10

11 Map Orientation and Projections 11

12 Map Orientation and Projections 12

13 Map Orientation and Projections Original Correction 13

14 Why? Why visualization? Get insights Come up with hypotheses Detect the expected, and discover the unexpected

15 Why? Why visualization? Get insights Come up with hypotheses Detect the expected, and discover the unexpected

16 Applications Mapping With all map applications throughout history Decision making E.g., disease outbreaks, crimes, etc Real-time monitoring E.g., traffic, security, etc Scientific analysis E.g., climate change, vegetation analysis, etc

17 Geo-visualization Element Three elements Data: what to visualize? Location: where to put data? Visualization scheme: how to visualize? 17

18 Geo-visualization Element Three elements Data: what to visualize? Location: where to put data? Visualization scheme: how to visualize? 18

19 Geo-visualization Element Three elements Data: what to visualize? Location: where to put data? Visualization scheme: how to visualize? 19

20 Interactive Maps MapD interactive demos Tweet map: 20

21 Interactive Maps MapD interactive demos US Census: 21

22 Interactive Maps Pan and Zoom (in interactive views) Pan: change your data focus on same spatial view level Zoom: change your spatial view level 22

23 Interactive Maps Pan and Zoom (in interactive views) Pan: change your data focus on same spatial view level Zoom: change your spatial view level Linking and Brushing (in multiple views) Linking: highlight certain part of data in all views Brushing: dynamic linking (linking + panning) This happens when you have multiple distinct views, e.g., a map, a table, and a graph, or a set of temporally partitioned views 23

24 Interactive Maps Pan and Zoom (in interactive views) Pan: change your data focus on same spatial view level Zoom: change your spatial view level Linking and Brushing (in multiple views) Linking: highlight certain part of data in all views Brushing: dynamic linking (linking + panning) This happens when you have multiple distinct views, e.g., a map, a table, and a graph, or a set of temporally partitioned views Specification of interactive visualization 200 ms response time (controversial) 24

25 Visualization in Virtual Reality

26 Big Spatial Data Visualization New challenges come with big volume data How to put data on the map? How to aggregate large data? How to process large data? 26

27 Big Spatial Data Visualization New challenges come with big volume data How to put data on the map? How to aggregate large data? How to process large data? High velocity High velocity data visualization exploits pre-materialization Still active research is on-going 27

28 Designing an Effective Visualization Need to take human perception into account 28

29 Designing an Effective Visualization Need to take human perception into account 29

30 Designing an Effective Visualization Communicate the right message 30

31 Designing an Effective Visualization Consider conflicted entities 31

32 Designing an Effective Visualization Consider conflicted entities 32

33 Designing an Effective Visualization Human perception is sensitive to: Sizing Colors perception (color choice, clarity, etc) Conflicted entities (names, borders, etc) Values, e.g., population vs population density 33

34 Designing an Effective Visualization Human perception is sensitive to: Sizing Colors perception (color choice, clarity, etc) Conflicted entities (names, borders, etc) Values, e.g., population vs population density Visualization confusions might be caused by: Too many colors Inconsistent scales Wrong chart types (e.g., continuous chart on discrete data). 34

35 Credits Prof. Luc Anselin s lecture 35